Neuropathic pain is common among Veterans, substantially impeding their attempts to rehabilitate function. Numerous contributing mechanisms have been identified, but have not led to any new therapies. Initial observations show that cannabinoids may hold promise for new therapeutic approaches. There is growing recognition of the participation of endocannabinoids (ECs), which are endogenous agonists of cannabinoid receptors (CB1R), in the central regulation of pain by descending inhibition of sensory pathways (antinociception), although their site of action has not been determined. There is evidence that the EC system in the dorsal periaqueductal gray (dPAG), a key pain regulatory center, may contribute to antinociception. Promising preliminary data from our lab shows an upregulation of cannabinoid signaling in the dPAG in a subset of rats that did not develop neuropathic pain after nerve injury. Therefore, we propose to explore the dPAG as a potential locus for pain control, and test the overall hypothesis that activation of the EC system in the dPAG drives descending analgesic signaling that suppresses neuropathic pain. The dPAG has the potential to coordinate analgesic mechanisms with autonomic control since activation and EC-modulation of dPAG neurons increases sympathetic nerve activity and blood pressure. We have previously established a link between susceptibility to development of neuropathic pain and autonomic activation by showing that rats with elevated initial sympathetic tone do not develop hyperalgesia when subjected to spinal nerve ligation (SNL). Sympathoexcitation is typically not a desired side effect of EC-induced analgesia in the setting of chronic pain. We will therefore also explore specific mechanisms whereby ECs in the dPAG could selectively regulate autonomic activation versus analgesia. The program is organized into three Specific Aims. In Specific Aim 1, we will characterize anatomic and pharmacologic features of dPAG circuitry specific to antinociception and sympathoexcitation. These explorations will provide critical insights into dPAG circuitry at a novel level of detail and establish whether changes in dorsal horn sensory neuron (DHN) activation and sympathetic outflow show patterns of differential control. Antinociception will be defined by inhibition of DHN neuron activity while sympathoexcitation will be recorded as elevated sympathetic nerve activity. Agents will be microinjected into the dPAG that excite neurons and activate or attenuate cannabinoid system function. An understanding of the dPAG mechanisms that integrate antinociception with sympathoexcitation will provide opportunities to dissociate the desirable cannabinoid antinociception from unwanted cardiovascular effects. In Specific Aim 2, we will evaluate injury-induced changes in the expression of components of the dPAG cannabinoid system. Specific molecular consequences underlying dPAG contributions to pain will be characterized in rats showing a range of hyperalgesia after SNL. Gene and protein expression levels of components of the cannabinoid system in the dPAG will be correlated with the degree to which hyperalgesia develops in individual rats. These experiments will test the hypothesis that upregulated EC signaling in the dPAG prevents the development of hyperalgesia following nerve injury, suggesting a therapeutic potential of dPAG cannabinoids. Finally, in Specific Aim 3, we will identify the role of the dPAG cannabinoid system in the development of neuropathic pain. Our preliminary findings support an important and previously unrecognized role of dPAG cannabinoid signaling in the generation of neuropathic pain. Cannabinoid system function in the dPAG will be attenuated acutely by microinjection of a CB1R antagonist or chronically by RNA interference to downregulate CB1R in rats undergoing SNL, a model of neuropathic pain, and the effects correlated with levels of hyperalgesia. These experiments will test the hypothesis that loss of CB1R function in the dPAG contributes to the development of neuropathic pain.